This invention relates generally to the field of electronic weighing scales, and more particularly to a method for recalibrating such scales to compensate for the effects on the accuracy of the scales of variations in physical and/or environmental conditions under which the scales operate.
This application is related to copending application Ser. No. 08/165,151, filed Dec. 12, 1993, issued on Aug, 27, 1996 as U.S. Pat. No. 5,550,328, in the name of Gerald C. Freeman and Paul C. Talmadge, and assigned to the assignee of this application.
Since their introduction, electronic scales have become widely accepted in many weighing applications for a number of reasons, primarily the extreme degree of accuracy with which the scales can weigh articles, the wide range of weights the scales are capable of handling and the ease and convenience of digital display readout of the weight of an article. Electronic scales are now used almost exclusively in such high volume utility situations as mail, parcels, bulk food and dry goods sold by weight measure, air terminal baggage, and other situations where highly accurate weight is required on a repetitive basis with minimum recovery time between individual weighings.
In recent years, electronic scales have become almost the universal standard in connection with weighing mail and parcels, and it is in connection with this utility that the present invention was developed, although the utility of the present invention is by no means limited to this use. Perhaps the primary contributing factor to the popularity of electronic scales in the postal field is the high degree of accuracy inherent in such scales. When one considers the billions of mail pieces weighed anually by the U.S. Postal Service in the course of handling mail, and the millions of packages and parcels also handled not only by the Postal Service but also by all of the special delivery courier services which compete with the Postal Service, one can begin to appreciate the vast amount of money, by which customers will be overcharged or undercharged depending on whether scales are overweighing or underweighing, in the course of dispatching all of this mail and parcels if the scales which determine the mail and parcel postage amount are not highly accurate.
For example, a generaly accepted standard of accuracy among major electronic scale manufacturers is that they be within 0.03% to 0.05% of full scale. If we assume a 100 pound scale, the required accuracy becomes 0.03 to 0.05 pounds, or 0.48 to 0.80 onces, over the range of the scale. Thus, it is apparent that electronic scales are capable of weighing accurately to an impressive less than one ounce in 100 pounds. Correspondingly, a one pound letter scale can weigh letter mail accurately to within less than one one hundreth of an ounce.
Aside from an inherent desire to provide highly accurate scales for monetary purposes described below, a major factor contributing to this high degree of accuracy is the requirement by the National Bureau of Weights and Measures that a scale be capable of weighing within the above limits of accuracy in order to be approved for commercial use in mail and parcel applications. Although many customers in other applications may not require this degree of accuracy, customers in the mail and parcel fields will not purchase scales that are not capable of National Bureau of Weights and Measures approval.
A major problem that occurs with electronic load cell scales is that the accuracy of the scales can be adversely affected by variations of certain physical and/or environmental conditions under which the scales are required to operate. A primary physical condition is that an electronic load cell scale must be absolutely level during operation or it will not weigh accurately. Thus, if a scale is properly calibrated at the factory on a test bench known to be perfectly level, and is then transported to the field and operated on a surface that is not as level as the factory test bench, the scale will not weigh accurately. Tests have revealed that a scale resting on a surface which is tipped only a few degrees off of factory test bench level can have a weight discrepancy of as much as 0.4% to 0.5% of full scale, which translates into an accuracy of about 10 times less than the above mentioned industry standard. The reason for this is that when a scale is perfectly level, an article resting on the platform of the scale is exerting 100% of its weight in a perfectly vertical direction relative to gravity, so the scale recognizes the full weight of the article. If the scale is tipped slightly, the weight of the article recognized by the scale is no longer 100%, but rather is only a component of the weight as determined by the cosine effect, one minus the cosine of the angle that the scale is off level. The result is that the scale reads less than the true weight of the article by the amount of the above percentages, which becomes very substantial in terms of lost revenue from underweighing millions of parcels. This problem could be particuarly acute in the situation where a courier service wishes to place a scale in the back of its pickup truck in order to check the accuracy of package weight provided by the customer prior to the package being delivered to the distribution center of the courier service. It is rare that a parked truck will be absolutely level, with the result that a substantial degree of error is introduced into the weight provided by the customer.
The problem is further compounded by the introduction of various environmental factors, such as variations in gravity, vibration, temperature, air movement, electronic noise, and shift errors on the platform. For example, it is known that the force of gravity varies from place to place around the world, with the result that a scale properly calibrated at the factory may not be accurate within the desired limits after it is transported a few hundred or a few thousand miles. Also, the effect of gravity varies with height, so that a scale calibrated properly at sea level may not be accurate within the desired limits in Denver. A scale properly calibrated in an air conditioned factory at a temperature of 75.degree. F. may not be accurate when used in a non-air conditioned location with an ambient temperature of 98.degree. F. Air movement is another contributing factor, so that a scale operating under the influence of air movement impinging on the platform from an air conditioning outlet may not be accurate after having been calibrated at the factory in still air.
Prior to the present invention, the only way to ensure that a particular scale would weigh accurately in the field was to dispatch a service technician of the scale manufacturer to the site of the scale for the purpose of recalibrating the scale after it is installed and is operating under the conditions which appertain. This obviously is not an acceptable solution since it is not a cost effective procedure, it does not solve the problem of changing circumstances at the scale site, such as temperature, air movement, electronic noise, etc., and it certainly does not solve the problem of transitory scales, such as those installed in the back of courier services' pickup trucks, and finally it is most difficult on the service technicians who must carry anywhere from 50 to 100 pounds of weight to the site of the scale in order to perform the calibration.
Thus, from the foreging it is seen that there exists a need for an effective way of recalibrating electronic scales in the field to compensate for errors in weight which are introduced by the adverse effects of physical and/or environmental conditions under which the scales operate.